Spatial Modeling of Industrial Windfall on Soils to Detect Woody Species with Potential for Bioremediation

S. Salazar, Colegio de Postgraduados, Campus , Carretera Federal Veracruz- km 26.5. Mpio. Manlio Fabio Altamirano, Veracruz, C. P. 91690 M. Mendoza, FIPRODEFO, Jalisco, México, A. L. del Angel Perez, Instituto Nacional de Investigaciones Agropecuarias, Campo Cotaxtla, Veracruz, Ver. A. Tejeda M, Facultad de Ciencias Atmosféricas, Universidad Veracruzana. Xalapa, Ver. México

Abstract—A spatial model is presented to explain the concentration of heavy metals (Fe, Cu, Zn, Ni, Cr, Co and Pb), in the soils around the industrial complex near the Port of Veracruz, . Unexpected low concentration sites where then tested to detect woody plant species that may have the capability to hiperacumulate these contaminants, hence having a potential for dendroremediation. The study case circumstances are conducive to test this hypothesis because of the radial nature of the industrial sprawl, and the rural surroundings, where low presence of heavy metals is expected. Contaminants selected are common in the sort of industrial outfits in the region, most of which are involved in metal manufacturing. Results show levels of Cu, Zn, Cr, Co, and Pb in soils exceeding the reference threshold concentration in a fanshaped pattern downwind, with peak values for Fe. At least four species had no damage symptoms: guácimo, Guazuma ulmifolia; roble, Tadelobuia pentaphylla L., cocuite, Gliricidia sepium and palo mulato, Bursela simaruba L. This suggests these materials are resistant to metal pollution at the concentration levels seen in the region. Analysis of woody and bark tissue is underway, and it is expected to confirm the potential of the four species as pollutant accumulators suitable for bioremediation. If such were the case, they should they be propagated and placed all through out the downwind side of the industrial complex in a large scale trial. The up wind concentrations were not a concern from the point of view of environmental regulations, and the effect of ac- cumulator species already present resulted not statistically significant, except cocuite.

Introduction Nowadays there is a recent body of regulatory legisla- tion concerning toxic emissions into the atmosphere, and Spatial modeling of airborne contaminants on soils heavy metals are a matter of concern in this regard. Tools is an approach capable of identifying woody species to locate and spatially analyze hot spots are therefore with potential capability for remediation and reclaiming. urgently needed. Screening up locations where con- There are two direct benefits from this type of result, a) taminants represent a hazard to population is an activity assess the soil health condition, and b) identify native of major priority. In a similar fashion, modeling the species suitable for clean up. fallout patterns creates the opportunity to detect species In the Mexican scene, 21 out of 32 states comprises with potential capability to accumulate these metallic 166 illegal toxic dump sites abandoned. An assessment contaminants. in 1995-1997 reported: heavy metals, dust, smelter The municipality of Veracruz (latitude 19º 16´ y 19º residues (scum), mining debris, oils, hydrocarbons, or- 06´ N, longitude 96º 07´ y 96º 20´ W) includes the city ganoclorides, muds, Pb, cadmium, niquel, cianure, silic of Veracruz (pop. 457 377), and a considerable industrial sand (EMA 1999). Consider in addition that industry infrastructure including the Port of Veracruz, and Bruno continuously disposes this and other contaminants. Pagliai industrial park (FIBP), (INEGI 2000).

USDA Forest Service Proceedings RMRS-P-42CD. 2006. 871 Climate in the study region is tropical semidry, with of bark and wood was taken at 1.3 m in heigth. Two soil summer rains. Weather is strongly influenced by trade samples from the tree crown cover area were also taken, winds. Natural vegetation cover in Veracruz has declined one at 10 cm in depth, and another one at 20 cm. For each historically, temperate forests dropped from 18 percent soil sample, total concentration of Fe, Cu, Zn, Mn, Ni, to 4 percent, and tropical forests from 66 percent to 24 Cr, Co, Cd, Pb, and extractable concentration of Fe, Cu, percent, in the 1800 to 1900 time span (INEGI 2000). Zn and Mn were determined with procedures indicated The main cover type is low tropical forest, which in- in the official Mexican norm about sampling and analysis volves wooded areas of up to 15 m in height, growing of soils NOM-021-RECNAT-2000 (SEMARNAT 2002). in a climate with 25.3 ºC mean temperature, and 1669.2 Tissue analysis was done through atomic absorption us- mm annual rainfall (SEFIPLAN 2002). This vegetation ing a solution of ashes from tissue sample incinerated responds to xeric site conditions imposed by the sandy, at 550 ºC. poorly developed soils usually found in coastal dunes Complementary soil data sought included pH, electric such as in the study region. Local species are well adapted conductivity, cation exchange capacity organic mater, to these environments produced by the fast drainage, total nitrogen, assimilable phosphorous, exchangeable occasional gusty winds, heavy showers, and high salt bases, and texture. content in the water. Climate data was provided by Centro de Previsión del This research has an appeal and relevance given by the Golfo de México. Wind speed and direction, temperature, evolving conditions of reduced tree cover, loss of natural relative humidity, global solar radiation, cloud cover, and landscape, increasing population and industrial develop- barometric pressure were variables selected to model ment, and rising emission levels (EMV 1998). atmospheric dispersal. The local cartography was digi- tized, and fed to software Archie (FEMA/USEPA 1993), and Surfer (Golden Software, Inc). Isoconcentration Materials and Methods digital maps for each contaminant were produced with the ground and atmosphere modeled features, plus con- The study area is defined by a 5 km radius centered at taminant values recorded in field samples of soil, bark FIBP, at 16 km west from the city and port of Veracruz. and wood. Scenarios of interest to model included: FIBP began with the first facility, Tenaris Tamsa, open- • the annual cycle, represented by weather and con- ing in 1958. Tenaris is a steel mill and pipe manufacturer taminant fallout patterns in January, April, July, and that is the largest (185 ha) and dominant tenant in the October (featuring seasonal conditions for winter, park. Today there are 157 industrial outfits in FIBP, 48 spring, summer, and fall), gusty wind days (synoptic of them are metal-mechanic manufacturers, and 44 are weather systems from higher latitudes descending chemical industries. Other businesses in FIBP involve on to the tropics; also, stormy conditions associated food and beverages, electric and electronic components, to anticyclonic circulation producing winds over 30 glass, construction materials, and some other products m s-1) (CFBP 2002). • steady days with high temperature (maximum hourly The federal highway Veracruz-Jalapa divides by half temperatures each month) the study area. This road is used by over 2000 vehicles, many of which are heavy trucks. Industrial facilities are surrounded by farms with Results cropping and grazing plots, though these activities are declining because of new housing developments. Soils presented an average pH of 5.95 at the 0-10 cm The largest of these housing projects is Valente Díaz depth, and 5.79 at 10-20 cm (±0.69 and ±0.79 standard (pop. 13765), which is the second largest population in deviation, respectively). Mean electric conductivity was Veracruz municipality, just surpassed by Veracruz city 0.142 dS.m-1, a value within acceptable range (United itself. One more factor of relevance for Valente Díaz is States Salinity Laboratory of Riverside, USA). Organic that it borders with the Tenaris steel mill (SEFIPLAN mater contents in the soil was >3.01 percent at 10-20 2002). cm deep, and still very high at 2.01 percent to 3.0 per- A network of sampling points were subjectively cho- cent in the top layer, at 0 cm to10 cm. Soils were coarse sen to supply information from different distances and in texture, ranging from sandy loam to loamy sand. orientation respect the center point of the study area. Exchangeable cations were also high, ranking Ca >Mg Sampling points necessarily must be near a native tree >K>Na. of sufficient age as to expect to have received the effect Soil total heavy metals (Fe, Cu, Zn, Mn, Ni, Cr, Co, of pollutants and reacted to it. From this tree, a sample Pb), resulted over the reference tolerance (table 1). Of

872 USDA Forest Service Proceedings RMRS-P-42CD. 2006. Table 1. Average of total heavy metals in soils 0-10 cm, 10-20 cm, mg.kg-1 (dry weight).

0-10cm 10-20cm

Fe This study This study (1) (2) (3) (4) average 45372.9 40411.7 - - - - range 7520-161590 20740-116550 Cu average 67.3 78.87 100 60 26 00 range .7-468.8 -437.5 Zn average .8 163.7 400 70 73. 00 range -1550 38-1240 Mn average 036 957. 00 000 490 - range 74-3950 298-3831.5 Ni average 42.9 43.16 100 - 8. 00 range 20-234 20-204.8 Cr average 2.3 105.7 75 - 0 00 range 18-675 18-675 Co average .3 35.3 2 0 0.5 - range .7-63. .7-63.5 Cd average 2 - 8 - -  range 0-2 - Pb average 05.2 111.9 200 - 26 00 range 26-250 20-440 particular concer was total and extractable Fe, where Table 2. Trees sampled in the study site. FIBP, Veracruz, Mexico. records show elevated concentrations >25.0 mg.kg-1, while Cu averaged a high >1.2 mg.kg-1, Zn was also Nombre común Nombre científico* beyond toxic levels (>8.0 mg.kg-1), and so was Mn too Almendro Terminalia catappa L. (>29 mg.kg-1) Cedro Cedrela odorata A total of 13 tree species were tested (table 2). Cocuite Gliricidia sepium Concentrations of Fe, Cu, Zn, Mn, Ni, Cr, Co, and Pb Ciruelo Spondias mombin L. Flamboyán Delenox regia in bark and wood tissues are presented at table 3. Guácimo Guazuma ulmifolia Cu bioacumulation index by species was oustanding, Higuera Ficus carica and statistically significant for cocuite (Gliricidia sepium) Mulato Bursela simaruba L. Jobo Spondias purpureas and roble (Tadelobuia pentaphylla L). Three other species Nacastle Enterolobium cyclocarpum also displayed a satisfactory bioacumulation potential Palo dulce Achatacarpus nigricans (index>1). Roble Tadelobuia pentaphylla L. Uvero Coccoloba barbadensis Discussion (Martínez 1994)

This first evaluation assessed heavy metals in the study area soils. The key finding is the expected behavior of buildings), will make contaminant drift even faster. In sandy and loamy soils. Coarse soils like these have a low this scenario, current environmental regulations might be retention of heavy metals. Therefore, these contaminants excessively lenient in protecting public health. seep through down into the water table. The ample empty Should any of the species studied turn out as good spaces inside sandy soils provide fast movement of air bioaccumulators, they could be a valuable means to and water, and therefore, quick movement of metals car- clean up medium level brown soils. Moreover, the spatial ried as particles or dissolved. model used to identify bioaccumulators is also capable of Now considering the current trend towards industrial identify priority areas to be forested with those species and housing development in the region, significant in- found, and the minimum and maximum area and planting crease of paved ground and artificial structures (houses, density can be calculated too.

USDA Forest Service Proceedings RMRS-P-42CD. 2006. 873 Table 3. Heavy metals in bark and wood, study site, Veracruz, México. mg.kg-1 (dry weight).

Bark Wood

Fe Cu Zn Mn Ni Cr Co Pb Fe Cu Zn Mn

Roble 407 28.7 26 6. . . 7.8 6.8 6.25 8.68 8.7 Cocuite 2 7.  23.7  8 6.2 23.7 Jobo 486 8.8  4 . 8 6. .2 2 8.6 7.9 Ciruelo 408 28.  7 . . .2 .6 6.25 6.2 Nacastle 677 2.5 24.8 20 . 8 .6 2.7 6.2 8.6 43.15 Nopillo 475 24.2 29.5 89.2 . 4.6 6. .36 8 6.2 2.4 24.92 Mulato  20.9 29. 06 . 6 .36 8 6.2 .95 25.32 Almendro 207 8.8 24.8 . . . .2 28.6 6.2 2.4 Palo Dulce 2 2.5 49.6 .8 . .2 0. 2.4 Guácimo 168 15.6 37.2 27.7 5.9 5.2 39.3 6.25 12.4 38.4 Uvero 4.7 6.2 2.8 82.4 4.2 6.25 8.4 .7 Cedro 78.7 6.2 2.4 2.4 48.2 6.2 2.4 68.6 Higuera 68.5 6.25 12.8 79.8 36.2 6.25 8.78 58.4 Flamboyán 67.3 6.2 2.7 84. . .5 6.2 .86 60.23

References Kovalskiy, V.V. 1974. Geochemical environment, health and diseases, in Trace Subst. Environ. Health, Vol. 8, Hemphill, D.D., Ed., University of Missouri, Columbia, Mo. 137p. CFBP 2002. Censo del Fideicomiso Industrial Bruno Pagliai, Linzon, S.N., 1978. Phytotoxicology Excessive Levels for 2002. Contaminants in Soil and Vegetation, report of Ministry of EMA, 1999,. Instituto Nacional de Estadística Geografía e the Environment, Ontario, Canada, 1978. Informática. Estadísticas del Medio Ambiente. Tomo II. Martínez, M. 1994 .catalogo de nombres vulgares y científicos México. de plantas mexicanas. Fondo de Cultura económico. EMV, 1998. Enciclopedia Municipal Veracruzana, Veracruz. 1248 p. Gobierno del Estado de Veracruz. Secretaría Técnica. SEFIPLAN, 2002. Secretaría de Finanzas y Planeación del 39-40 p. Gobierno del Estado de Veracruz. FEMA/USEPA, 1993. Automated Resource for Chemical SEMARNAT, 2002. Secretaría de Medio Ambiente y Recursos Hazard Incident Evaluation (ARCHIE) Federal Emergency Naturales, Norma Oficial Mexicana NOM-021-RECNAT- Management Agency/United States Environmental 2000, Fertilidad, salinidad y clasificación de suelos. Protection. Risk Managent Program. Estudios, muestreo y análisis INEGI, 2000. Instituto Nacional de Estadística, Geografía e (1) Linzon, 1978 Informática, Municipio de Veracruz. Kabata-Pendias, A, and H. Pendias. 1992. Trace Elements (2) Kovalskiy, 1974 in Soils and Plants. CRC Press, Inc. Boca Raton. Florida. (3) USEPA, 1995 11-55p. (4) Kabata-Pendias and Pendias, 1984.

874 USDA Forest Service Proceedings RMRS-P-42CD. 2006.